The E4 allele of the Apolipoprotein E (ApoE) gene is the strongest genetic risk factor for the onset of sporadic Alzheimer's disease (AD) identified to date. The roles by which ApoE influences amyloid-beta (A2) metabolism and non-A2-mediated mechanisms in AD pathogenesis, however, remain to be fully clarified. The literature, and our preliminary data, suggest that early alterations in peripheral lipids (sphingolipids, fatty acids, cholesterol and cholesterol esters) reflect brain functioning and pathology, and may interact with ApoE genotype in the development of AD pathogenesis, warranting human studies. Clinical and epidemiological studies of short duration, while important, will not contribute to our understanding of the earliest phases of AD pathogenesis because Alzheimer's pathology (A2 plaques and neurofibrillary tangles) begins decades before the emergence of symptoms and substantial neurodegeneration. Identifying factors years before the onset of AD that may modify the effects of ApoE4 in initiating and promoting AD pathology and the subsequent emergence of symptoms will uniquely contribute to the development of prevention strategies. Longitudinal studies of cognitively normal individuals with serial measures of Alzheimer's pathology in the living brain, as proposed here in the unique cohort of the Baltimore Longitudinal Study of Aging (BLSA), initiated in 1958, are necessary to understand the relationship between ApoE4 genotype, perturbations in peripheral lipids, their interaction, and later development of AD clinical symptoms and brain alterations. The BLSA is one of few human studies that could provide the unprecedented opportunity to systematically examine this relationship over a long follow-up. BLSA participants, cognitively normal at their first visit in the study (mean age: 63.4), have a mean follow-up of 14.3 years (SD = 6.5) and a maximum follow-up of 38.9 years. In the proposed study we will measure plasma lipid levels (sphingolipids, fatty acids, cholesterol and cholesterol esters) at three early visit in the BLSA study, roughly 5 years apart, for those aged 55 and over (n=1095), and at the last visit, as well as during the neuroimaging sub-study. The specific aims include examining the proposed peripheral lipids, changes in these lipids over a long follow-up, and their interaction with ApoE to predict: 1) decline in tests of memory; 2) incident MCI, all-cause dementia, and AD; 3) change in serial MRI measures of brain atrophy and white matter lesion burden over 10 years; and 4) amyloid-beta deposition on 11C-PIB PET scans. The lipids will be assayed using an already-developed targeted and quantitative lipidomic approach.